For genomic analyses of bacterial genomes, horizontal gene transfer is a complicating factor, causing different genes to have discordant histories. I plan to develop methods for estimating bacterial species phylogeny in the face of this horizontal gene transfer. I will also develop methods that allow the biologist to uncover patterns in the data, by grouping together genes that have similar realized values of evolutionary parameters. Finally, I will develop more realistic models of protein evolution that take advantage of the most cutting-edge work for the ab initio prediction of protein structure. Using the same statistical framework, I will attempt to predict RNA secondary structure by combining information on the Gibb's free energy with comparative sequence information. The methods developed in the course of this research will be implemented in the next generation of the MrBayes software. 1 PUBLIC HEALTH RELEVANCE: The comparison of DNA sequences from different organisms allows scientists to learn about the processes that shape genomes. For example, by comparing the DNA sequences of deferent genes, we can learn about how natural selection has acted. This proposal develops improved methods for the analysis of DNA sequences. 1